Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Prediction Algorithms”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 145 records · Page 8Linked to original sources
PubMed2002

A blind comparison of four non-invasive twelve-lead electrocardiogram algorithms for predicting susceptibility to ventricular tachycardia.

INTRODUCTION: To reduce QT measurement error, a new method was tested in which high-gain, high-speed, simultaneous 12-lead electrocardiographic (ECG) recordings were obtained during a single cardiac cycle. To increase its predictive power, the utility of combining QTD with the QRS duration for predicting susceptibility to ventricular tachyarrhythmia (VT) was analyzed. METHODS AND RESULTS: A total of 113 patients referred for electrophysiological study underwent baseline simultaneous 12-lead ECG followed by electrophysiological study to determine VT inducibility. Twenty-six patients had inducible VT while 87 patients did not. QT intervals and the width of QRS complex were measured from a single cardiac cycle with high-gain (8 times normal) and high-speed (100 mm/second) 12-lead ECG recordings. This method resulted in 100% QT interval identification throughout all 12 leads for every patient. Receiver-operator characteristic curves (ROC) and the areas under the ROC curves (AUC) were used to quantitatively analyze the performance of four ECG variables (QTD3, QTD12, QTD12 + QRS and QTD3 + QRS). All four ECG variables were significantly increased in the patients with inducible VT as compared to those without inducible VT. The QTD3 algorithm was less useful than QTD12 in predicting inducible VT; however, the addition of QRS duration to all QTD algorithms enhanced VT detection. CONCLUSION: 1) QRS duration has an incremental benefit in the detection of VT when combined with QTD; 2) QTD12 + QRS duration provided the highest predictive power among the four tested algorithms; 3) high-gain, high-speed 12-lead ECG recordings reduced QT measurement error.

Aged↗
PubMed2004

A comprehensive comparison of comparative RNA structure prediction approaches.

BACKGROUND: An increasing number of researchers have released novel RNA structure analysis and prediction algorithms for comparative approaches to structure prediction. Yet, independent benchmarking of these algorithms is rarely performed as is now common practice for protein-folding, gene-finding and multiple-sequence-alignment algorithms. RESULTS: Here we evaluate a number of RNA folding algorithms using reliable RNA data-sets and compare their relative performance. CONCLUSIONS: We conclude that comparative data can enhance structure prediction but structure-prediction-algorithms vary widely in terms of both sensitivity and selectivity across different lengths and homologies. Furthermore, we outline some directions for future research.

Algorithms↗
PubMed1990

Topology of receptor binding domains of mouse IFN-gamma.

IFN-gamma is an essential immunoregulatory lymphokine for a variety of immunologic functions including upregulation of MHC Ag. The elucidation of the structure, particularly the receptor binding domains, should further enhance our understanding of its mechanism of action, and provide a rational basis for modulation of its activity by alteration of its structure. A predicted model of murine IFN-gamma structure has been constructed based on data derived from our synthetic peptide studies, circular dichroism spectra, and predictive algorithms for secondary structure, surface accessibility, and tertiary structure, as well as predicted structural similarities to IL-2. Direct synthetic peptide competition studies using five overlapping peptides that encompassed the entire IFN-gamma sequence of 133 amino acids showed that only the N-terminus of IFN-gamma, IFN-gamma(1-39), blocked binding to receptor, suggesting that the N-terminus is directly involved in receptor binding. Rabbit antibodies to the N-terminal (IFN-gamma(1-39)) and C-terminal (IFN-gamma(95-133)) peptides neutralized IFN-gamma activity, whereas antibodies to the three peptides that spanned the internal region, sequence 36-107, were without effect. Thus, the antibody data support the involvement of the N-terminus as a receptor binding domain and also suggest that the C-terminus of the molecule is also a binding domain. Predictive algorithms assign six alpha-helices and five turns to the molecule and circular dichroism spectra of both intact human and murine IFN-gamma and synthetic peptides of murine IFN-gamma showed that the molecule is mainly alpha-helical in structure. Drawing mainly on the four alpha-helix bundle model, a common motif in globular proteins such as IFN-gamma and IL-2 (whose crystalline structure is known), we constructed a simple model of the tertiary structure of IFN-gamma that fits well with our synthetic peptide and circular dichroism data. The model consists of the four-alpha-helical bundle along with N- and C-terminal helices that are predicted to be closely associated and together form the receptor binding domains. The model presented should contribute to further understanding the molecular basis of IFN-gamma action and allow us to begin modulating the function of IFN-gamma through the design of agonists and antagonists.

Amino Acid Sequence↗
PubMed2002

Comparative analysis of human genome assemblies reveals genome-level differences.

Previous comparative analysis has revealed a significant disparity between the predicted gene sets produced by the International Human Genome Sequencing Consortium (HGSC) and Celera Genomics. To determine whether the source of this discrepancy was due to underlying differences in the genomic sequences or different gene prediction methodologies, we analyzed both genome assemblies in parallel. Using the GENSCAN gene prediction algorithm, we generated predicted transcriptomes that could be directly compared. BLAST-based comparisons revealed a 20-30% difference between the transcriptomes. Further differences between the two genomes were revealed with protein domain PFAM analyses. These results suggest that fundamental differences between the two genome assemblies are likely responsible for a significant portion of the discrepancy between the transcript sets predicted by the two groups.

Chromosome Mapping↗
PubMed1999

Defining the transmembrane helix of M2 protein from influenza A by molecular dynamics simulations in a lipid bilayer.

Integral membrane proteins containing at least one transmembrane (TM) alpha-helix are believed to account for between 20% and 30% of most genomes. There are several algorithms that accurately predict the number and position of TM helices within a membrane protein sequence. However, these methods tend to disagree over the beginning and end residues of TM helices, posing problems for subsequent modeling and simulation studies. Molecular dynamics (MD) simulations in an explicit lipid and water environment are used to help define the TM helix of the M2 protein from influenza A virus. Based on a comparison of the results of five different secondary structure prediction algorithms, three different helix lengths (an 18mer, a 26mer, and a 34mer) were simulated. Each simulation system contained 127 POPC molecules plus approximately 3500-4700 waters, giving a total of approximately 18,000-21,000 atoms. Two simulations, each of 2 ns duration, were run for the 18mer and 26mer, and five separate simulations were run for the 34mer, using different starting models generated by restrained in vacuo MD simulations. The total simulation time amounted to 11 ns. Analysis of the time-dependent secondary structure of the TM segments was used to define the regions that adopted a stable alpha-helical conformation throughout the simulation. This analysis indicates a core TM region of approximately 20 residues (from residue 22 to residue 43) that remained in an alpha-helical conformation. Analysis of atomic density profiles suggested that the 18mer helix revealed a local perturbation of the lipid bilayer. Polar side chains on either side of this region form relatively long-lived H-bonds to lipid headgroups and water molecules.

Amino Acid Sequence↗
PubMed1995

Characterization of four new tcp-1-related cct genes from the nematode Caenorhabditis elegans.

In this report we present the sequences of four new cct chaperonin genes from the nematode Caenorhabditis elegans. The four genes, cct-2, cct-4, cct-5, and cct-6 are orthologs of the mouse chaperonin genes Cctb, Cctd, Ccte, and Cctz, sharing 66%, 63%, 68%, and 67% deduced amino acid sequence identity, respectively. The C. elegans multigene family includes these four genes as well as cct-1 (tcp-1), and displays 23-35% pairwise predicted amino acid sequence identity between members, and 31-35% identity to the closely related archaebacterial chaperonin TF55. The five C. elegans cct genes are expressed in all life stages (egg, four larval stages, and adult). Members of the multigene family occur as a loosely associated group of three genes on chromosome II, and two widely separated genes on chromosome III. The predicted secondary structures of all five C. elegans CCT deduced protein sequences are nearly identical. Moreover, all chaperonins examined had comparable predicted secondary structures. Algorithmic predictions of the secondary structures of GroEL, Hsp60, and Rubisco subunit-binding protein (RuBP) are almost identical, and are very similar to the known GroEL secondary structure. The CCT/TF55 family predicted secondary structures are essentially identical to each other and are also related to GroEL, Hsp60, and RuBP. The most notable difference between the CCT/TF55 and the GroEL/Hsp60/RuBP families is in the presumed polypeptide binding domain.

Amino Acid Sequence↗
PubMed2006

Localizome: a server for identifying transmembrane topologies and TM helices of eukaryotic proteins utilizing domain information.

The Localizome server predicts the transmembrane (TM) helix number and TM topology of a user-supplied eukaryotic protein and presents the result as an intuitive graphic representation. It utilizes hmmpfam to detect the presence of Pfam domains and a prediction algorithm, Phobius, to predict the TM helices. The results are combined and checked against the TM topology rules stored in a protein domain database called LocaloDom. LocaloDom is a curated database that contains TM topologies and TM helix numbers of known protein domains. It was constructed from Pfam domains combined with Swiss-Prot annotations and Phobius predictions. The Localizome server corrects the combined results of the user sequence to conform to the rules stored in LocaloDom. Compared with other programs, this server showed the highest accuracy for TM topology prediction: for soluble proteins, the accuracy and coverage were 99 and 75%, respectively, while for TM protein domain regions, they were 96 and 68%, respectively. With a graphical representation of TM topology and TM helix positions with the domain units, the Localizome server is a highly accurate and comprehensive information source for subcellular localization for soluble proteins as well as membrane proteins. The Localizome server can be found at http://localizome.org/.

Cell Compartmentation↗
PubMed2005

PRISM: protein interactions by structural matching.

Prism (http://gordion.hpc.eng.ku.edu.tr/prism) is a website for protein interface analysis and prediction of putative protein-protein interactions. It is composed of a database holding protein interface structures derived from the Protein Data Bank (PDB). The server also includes summary information about related proteins and an interactive protein interface viewer. A list of putative protein-protein interactions obtained by running our prediction algorithm can also be accessed. These results are applied to a set of protein structures obtained from the PDB at the time of algorithm execution (January 2004). Users can browse through the non-redundant dataset of representative interfaces on which the prediction algorithm depends, retrieve the list of similar structures to these interfaces or see the results of interaction predictions for a particular protein. Another service provided is interactive prediction. This is done by running the algorithm for user input structures.

Algorithms↗
PubMed2000

Thermodynamic parameters for DNA sequences with dangling ends.

The thermodynamic contributions to duplex formation of all 32 possible single-nucleotide dangling ends on a Watson-Crick pair are reported. In most instances, dangling ends are stabilizing with free energy contributions ranging from +0.48 (GT(A)) to-0.96 kcal/mol (). In comparison, Watson-Crick nearest-neighbor increments range from -0. 58 (TA/AT) to -2.24 (GC/CG) kcal/mol. Hence, in some cases, a dangling end contributes as much to duplex stability as a Watson-Crick A-T base pair. The implications of these results for DNA probe design are discussed. Analysis of the sequence dependence of dangling-end stabilities show that the nature of the closing base pair largely determines the stabilization. For a given closing base pair, however, adenine dangling ends are always more or equally as stable as the other dangling nucleotides. Moreover, 5' dangling ends are more or equally as stabilizing as their 3' counterparts. Comparison of DNA with RNA dangling-end motifs shows that DNA motifs with 5' dangling ends contribute to stability equally or more than their RNA counterparts. Conversely, RNA 3' dangling ends contribute to stability equally or more than their DNA counterparts. This data set has been incorporated into a DNA secondary structure prediction algorithm (DNA MFOLD) (http://mfold2.wustl.edu/mfold/dna/for m1.cgi) as well as a DNA hybridization prediction algorithm (HYTHERtrade mark) (http://jsl1.chem.wayne.edu/Hyther/hythermenu .html).

Base Pairing↗
PubMed2003

Scoring algorithm to predict survival after nephrectomy and immunotherapy in patients with metastatic renal cell carcinoma: a stratification tool for prospective clinical trials.

BACKGROUND: The objective of this study was to develop an algorithm capable of stratifying the survival of patients with metastatic renal cell carcinoma (RCC) after nephrectomy and immunotherapy. METHODS: The medical records of 173 patients who underwent radical nephrectomy for metastatic RCC and received recombinant interleukin-2 (IL-2)-based immunotherapy between 1989 and 2000 were evaluated. Survival was the primary endpoint and was assessed based on clinical, surgical, and pathologic parameters. The clinical parameters included age, gender, performance status, existing hypertension, thyroid-stimulating hormone (TSH) levels, location of metastases, and presenting symptomatology. The surgical features included the requirement for blood transfusion or adrenalectomy. The pathologic factors involved tumor stage, tumor size, nuclear grade, lymph node status, and histologic subtype. Disease-specific survival was estimated using the Kaplan-Meier method. Univariate and multivariate Cox proportional hazards models were used to determine associations between clinical and pathologic features and survival. RESULTS: The median follow-up was 3.2 years (range, 0.2-9.3 years). Death due to RCC occurred in 123 patients (71%) at a median of 13 months (range, from 0.1 months to 8.4 years) after nephrectomy. Multivariate analysis revealed that the following features were associated with survival: lymph node status (P = 0.002), constitutional symptoms (P = 0.005), location of metastases (P < 0.001), sarcomatoid histology (P = 0.003), and TSH level (P = 0.038). A scoring system based on the features in the multivariate model was created to stratify patients into low-risk, intermediate-risk, and high-risk groups. Estimated survival rates at 1 years, 3 years, and 5 years were 92%, 61%, and 41%, respectively, for the low-risk group and 66%, 31%, and 19%, respectively, for the intermediate risk group. The high-risk group had 1% survival at 1 year and no survivors at 3 years. CONCLUSIONS: In patients with metastatic RCC who were treated with nephrectomy and IL-2 immunotherapy, regional lymph node status, constitutional symptoms, location of metastases, sarcomatoid histology, and TSH levels were associated with survival. The authors present a scoring algorithm based on these features that can be used to predict survival in patients who present with metastatic RCC and to stratify such patients for prospective clinical trials.

Adult↗
PubMed2003

Nearest neighbour algorithm for predicting protein subcellular location by combining functional domain composition and pseudo-amino acid composition.

In this paper, based on the approach by combining the "functional domain composition" [K.C. Chou, Y. D. Cai, J. Biol. Chem. 277 (2002) 45765] and the pseudo-amino acid composition [K.C. Chou, Proteins Struct. Funct. Genet. 43 (2001) 246; Correction Proteins Struct. Funct. Genet. 2044 (2001) 2060], the Nearest Neighbour Algorithm (NNA) was developed for predicting the protein subcellular location. Very high success rates were observed, suggesting that such a hybrid approach may become a useful high-throughput tool in the area of bioinformatics and proteomics.

Algorithms↗
PubMed2026

Identification of MHC Ligands Through Allele-Guided Isolation Combined With Machine Learning for Improved MHC Assignment Using ARDisplay-I.

The isolation of major histocompatibility complex (MHC) ligands and subsequent analysis by mass spectrometry is considered the gold standard for defining targets for T cell-based immunotherapies. However, as many targets of high tumor specificity are only presented at low abundance on the cell surface of tumor cells, the efficient isolation of these peptides is crucial for their successful detection. Here, we demonstrate how optimizing the MHC ligand isolation strategy, based on both the presenting MHC alleles and the individual peptide level, enhances the identification of specific MHC ligands. This ideally acknowledges not only the hydrophobicity but also the post-translational modifications of the respective MHC ligands. To further improve the identification and characterization of MHC ligands, we developed an MHC class I ligand prediction algorithm (ARDisplay-I) that outperforms current state-of-the-art tools when benchmarked against competitors such as netMHCpan 4.1, MixMHCpred, or MHCflurry. Implementing these strategies can augment the development of T cell receptor-based therapies by improving the identification of novel immunotherapy targets and enriching the resources available in the computational immunology field through a superior MHC presentation prediction algorithm.

Ligands↗
PubMed1999

A behavior-based inverse kinematics algorithm to predict arm prehension postures for computer-aided ergonomic evaluation.

In this paper, the computational problem of inverse kinematics of arm prehension movements was investigated. How motions of each joint involved in arm movements can be used to control the end-effector (hand) position and orientation was first examined. It is shown that the inverse kinematics problem due to the kinematic redundancy in joint space is ill-posed only at the control of hand orientation but not at the control of hand position. Based upon this analysis, a previously proposed inverse kinematics algorithm (Wang et Verriest, 1998a) to predict arm reach postures was extended to a seven-DOF arm model to predict arm prehension postures using a separate control of hand position and orientation. The algorithm can be either in rule-based form or by optimization through appropriate choice of weight coefficients. Compared to the algebraic inverse kinematics algorithm, the proposed algorithm can handle the non-linearity of joint limits in a straightforward way. In addition, no matrix inverse calculation is needed, thus avoiding the stability and convergence problems often occurring near a singularity of the Jacobian. Since an end-effector motion-oriented method is used to describe joint movements, observed behaviors of arm movements can be easily implemented in the algorithm. The proposed algorithm provides a general frame for arm postural control and can be used as an efficient postural manipulation tool for computer-aided ergonomic evaluation.

Adult↗
PubMed1998

Antipredictable sequences: harder to predict than random sequences.

For any discrete-state sequence prediction algorithm A, it is always possible, using an algorithm B no more complicated than A, to generate a sequence for which A's prediction is always wrong. For any prediction algorithm A and sequence x, there exists a sequence y no more complicated than x, such that if A performs better than random on x, then it will perform worse than random on y by the same margin. An example of a simple neural network predicting a bit sequence is used to illustrate this very general but not widely recognized phenomenon. This implies that any predictor with good performance must rely on some (usually implicitly) assumed prior distributions of the problem.

Algorithms↗
PubMed2005

Evaluation of capnography using a genetic algorithm to predict PaCO2.

INTRODUCTION: Noninvasive estimates of Paco(2) are usually done by measuring exhaled carbon dioxide at end-expiration (Petco(2)). While commonly used in studies involving healthy patients, it is less useful in sicker patients. Conditions that affect the terminal dead space and hence the accuracy of Petco(2) as a surrogate for Paco(2) may also affect other components of the capnogram. A genetic algorithm is a computer technique for discovering relationships between variables. The purpose of this study was to use a genetic algorithm to improve the precision of Paco(2) prediction in comparison to Petco(2). METHODS: Inspiratory and expiratory volumes were measured and analyzed by the computerized capnogram. Data were recorded for 2 min. Within 5 min of recording the capnograms, arterial blood gases were obtained. After excluding artifact and incomplete capnograms, five of the remaining breaths from each patient were selected. A genetic algorithm, constructed in postfix notation, consisted of 1,000 chromosomes with genes randomly selected from the 11 capnographic data fields and mathematical operators. The algorithm was constructed on 400 breaths from 83 randomly selected patients (construction group) and tested on 160 breaths from the remaining 32 patients (test group). RESULTS: For the construction group, the bias and precision between Petco(2) and Paco(2) were 4.3 +/- 4.9 mm Hg (mean +/- SD). For the 160 breaths in the test group, Petco(2) predicted Paco(2) with bias and precision of 2.9 +/- 4.2 mm Hg. The best chromosome found by the genetic algorithm was (10 x 5 + 5 x 5 x 5)/(10 x 10) x Petco(2) - (5 x 5 x 10 + 5 x 5)/(10 x 10) x int time + 2 x 2 x 2 x 2 + (2 x 2)/10, which reduces to 0.65 x Petco(2) - 2.75 x int time + 16.4. This produced a bias and precision of 0.9 +/- 4.1 mm Hg in the construction group and 0 +/- 3.7 mm Hg in the test group (p < 0.01). CONCLUSIONS: In this study of nonintubated emergency department patients, a genetic algorithm produced an improvement in bias and precision of Paco(2) prediction.

Adult↗
PubMed1997

Predicting time to nursing home care and death in individuals with Alzheimer disease.

OBJECTIVE: To develop and validate an approach that uses clinical features that can be determined in a standard patient visit to estimate the length of time before an individual patient with Alzheimer disease (AD) requires care equivalent to nursing home placement or dies. DESIGN: Prospective cohort study of 236 patients, followed up semiannually for up to 7 years. A second validation cohort of 105 patients was also followed. SETTING: Three AD research centers. PATIENTS: All patients met National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA) criteria for probable AD and had mild dementia at the initial visit. INTERVENTION: Predictive features, ascertained at the initial visit, were sex, duration of illness, age at onset, modified Mini-Mental State Examination (mMMS) score, and the presence or absence of extrapyramidal signs or psychotic features. MAIN OUTCOME MEASURES: (1) Requiring the equivalent of nursing home placement and (2) death. RESULTS: Prediction algorithms were constructed for the 2 outcomes based on Cox proportional hazard models. For each algorithm, a predictor index is calculated based on the status of each predictive feature at the initial visit. A table that specifies the number of months in which 25%, 50%, and 75% of patients with any specific predictor index value are likely to reach the end point is then consulted. Survival curves for time to need for care equivalent to nursing home placement and for time to death derived from the algorithms for selected predictor indexes fell within the 95% confidence bands of actual survival curves for patients. When the predictor variables from the initial visit for the validation cohort patients were entered into the algorithm, the predicted survival curves for time to death fell within the 95% confidence bands of actual survival curves for the patients. CONCLUSIONS: The prediction algorithms are a first but promising step toward providing specific prognoses to patients, families, and practitioners. This approach also has clear implications for the design and interpretation of clinical trials in patients with AD.

Age of Onset↗
PubMed2003

A prototype methodology combining surface-enhanced laser desorption/ionization protein chip technology and artificial neural network algorithms to predict the chemoresponsiveness of breast cancer cell lines exposed to Paclitaxel and Doxorubicin under in vitro conditions.

An ability to predict the likelihood of cellular response towards particular chemotherapeutic agents based upon protein expression patterns could facilitate the identification of biological molecules with previously undefined roles in the process of chemoresistance/chemosensitivity, and if robust enough these patterns might also be exploited towards the development of novel predictive assays. To ascertain whether proteomic based molecular profiling in conjunction with artificial neural network (ANN) algorithms could be applied towards the specific recognition of phenotypic patterns between either control or drug treated and chemosensitive or chemoresistant cellular populations, a combined approach involving MALDI-TOF matrix-assisted laser desorption/ionization-time of flight mass spectrometry, Ciphergen protein chip technology and ANN algorithms have been applied to specifically identify proteomic 'fingerprints' indicative of treatment regimen for chemosensitive (MCF-7, T47D) and chemoresistant (MCF-7/ADR) breast cancer cell lines following exposure to Doxorubicin or Paclitaxel. The results indicate that proteomic patterns can be identified by ANN algorithms to correctly assign 'class' for treatment regimen (e.g. control/drug treated or chemosensitive/chemoresistant) with a high degree of accuracy using boot-strap statistical validation techniques and that biomarker ion patterns indicative of response/non-response phenotypes are associated with MCF-7 and MCF-7/ADR cells exposed to Doxorubicin. We have also examined the predictive capability of this approach towards MCF-7 and T47D cells to ascertain whether prediction could be made based upon treatment regimen irrespective of cell lineage. Models were identified that could correctly assign class (control or Paclitaxel treatment) for 35/38 samples of an independent dataset. A similar level of predictive capability was also found (> 92%; n = 28) when proteomic patterns derived from the drug resistant cell line MCF-7/ADR were compared against those derived from MCF-7 and T47D as a model system of drug resistant and drug sensitive phenotypes. This approach might offer a potential methodology for predicting the biological behaviour of cancer cells towards particular chemotherapeutics and through protein isolation and sequence identification could result in the identification of biological molecules associated with chemosensitive/chemoresistance tumour phenotypes.

Algorithms↗
PubMed2004

Utility of MELD and Child-Turcotte-Pugh scores and the Canadian waitlisting algorithm in predicting short-term survival after liver transplant.

BACKGROUND: The Model for End-Stage Liver Disease (MELD) and Child-Turcotte-Pugh (CTP) scores are important predictors for survival after liver transplantation (LT). The objective of this study was to compare the utility of MELD and CTP scores with Canadian waitlisting algorithm in transplantation (CanWAIT) status for predicting 90-day survival after LT. METHODS: Retrospectively, we analyzed all 228 liver transplants performed in adults by the Atlantic Liver Transplant Program since 1985. These cases included combined transplants, retransplants and those after fulminant liver failure. MELD and CTP scores were calculated, and CanWAIT status and waiting time on the day of LT determined. We used c-statistic for 90-day outcome as the endpoint (survival), comparing areas under the receiver operating characteristic (ROC) curves for MELD and CTP scores and CanWAIT status. RESULTS: Mean (and standard deviation [SD]) MELD score was 18 (SD 12); CTP score, 10 (SD 3); and waiting time, 97 (SD 132) days. At the time of LT, 54% were in CanWAIT status 1; 4% in 1T; 14% in 2; 11% in 3; 6% in 3F; 4% in 4; and 7% in status 4F. Overall 90-day survival was 80% (95% confidence interval [CI] 75%-85%), exceeding the predicted survival by MELD scale with transplant of only 51% (CI 47%-55%). By c-statistic, CanWAIT is a clinically relevant predictor of 90-day outcomes in LT. By multivariate regression analysis, only CanWAIT status and age were found to have independent associations for short-term outcomes after LT. INTERPRETATION: CanWAIT status stratifies LT patients better and predicts short-term outcome more accurately than MELD or CTP scores, and so should not be replaced by MELD or CTP scores. This observation should be confirmed by a prospective and multicentre study in Canada.

Adult↗